Course Syllabus: CPSCI 240
Reasoning About Uncertainty – Fall 2013
Note: Scheduling and room assignments are tentative and reflect available information as of Aug
27, 2013. For the most recent scheduling information, please check Spire.
Section Times and Locations
Time:
Lectures
Discussion D01
Discussion D02
Discussion D03
M/W/F 2:30-3:20pm
F 1:25-2:15
F 10:10-11:00
12:20-1:10
LGRC A201
LGRC A201
LGRT 173
Location: Hasbrouck 134
Course Staff
Instructor
Teaching Assistant
Name:
Prof. Benjamin M. Marlin Yi Lu
Office:
CS234
Teaching Assistant
Kyle Wray
Office Hours: TBA
Email:
cs240staff@cs.umass.edu
Course Description: This course is designed to help you develop the mathematical reasoning
skills needed to solve problems involving uncertainty. The skills you will learn are crucial for
many exciting areas of computer science that inherently involve uncertainty, including artificial
intelligence and machine learning, data mining, financial modeling, natural language processing,
bioinformatics, web search and information retrieval, algorithm design, cryptography, system
design, network analysis, and more. These skills may also help you analyze the uncertainty in
your day-to-day life. The course is divided into three parts: sample space and probability;
random variables and expectations; and modeling, inference and estimation.
Prerequisites: CMPSCI 187 (or ECE 242) and MATH 132.
Textbook: Introduction to Probability, 2nd Edition. Dimitri P. Bertsekas and John N. Tsitsiklis.
Athena Scientific. 2008.
Course Website: The course website will be hosted on UMass's new Moodle course
management portal https://moodle.umass.edu/. You log into the portal using your OIT NetID and
password. The course website will host lecture notes, assignments, pointers to readings and
videos, announcements, and discussion forums.
Announcements: Official announcements for the course will go out through the Moodle portal
as email and will be automatically logged as news items. Email will be sent to your official
UMass email address (@student.umass.edu, @cns.umass.edu, etc...).
Grading Plan: The coursework will consist of homework assignments (written and
programming), discussion exercises, two midterm exams and a final exam. There is also a
participation component (in class participation, iClicker quizzes, asking and answering
discussion forum questions, etc.). The grading plan is given below:
Participation
5%
Discussion Exercises
10%
Homework Assignments
27%
Midterm Exam 1
15%
Midterm Exam 2
15%
Final Exam
28%
Course Policies:
•
Homework Submission: Written homework assignments must be submitted to the
CS240 drop box in the CS main office before 4:00pm on the day they are due.
Programming assignments must be uploaded to Moodle by the specified deadline.
Unlimited resubmissions are permitted on Moodle up to the deadline.
•
Late Homework: In general, late homework will not be accepted. In the case of an
extraordinary situation documented by suitable evidence (doctor’s note, etc...), an
extension or exemption will be granted.
•
Homework Collaboration: You are encouraged to discuss solution ideas for
assignments and other course material with other students. However, you must write
solutions to written problems and code for programming problems individually. Copying
is not permitted. Neither is collaboration so close that it looks like copying. If identical
homework solutions are detected, all will receive a grade of 0. A good practice is to
divide your work into an ideas phase where you collaborate, followed by a writeup or
coding phase where you work alone. You must write a list of all your collaborators at the
top of each assignment. This list should include anyone with whom you have discussed
the assignment.
•
External Resources: If you make use of any printed or online reference sources while
working on an assignment (other than specific course materials such as the textbook or
slides), these must be listed as references in your write-up or code. Copying written
solutions or code from the web is not permitted and is considered cheating.
•
Re-grading Policy: If you believe you've found a grading error, submit a detailed
written request along with your assignment or exam to the instructor (not an email)
explaining where you believe a mistake was made. Note that re-grading may result in
your original grade increasing or decreasing as appropriate.
•
Attendance: Students are expected to attend each class as well as the discussion section
they are registered for.
•
iClicker Use: Part of your participation grade will be based on answering in-class
iClicker questions. You must be physically present in class to participate. Having
someone else click for you constitutes cheating both by you and the person clicking
for you.
Approximate Schedule: (Subject to change over the semester)
Activity
Lecture 1
Lecture 2
Discussion 1
Lecture 3
Lecture 4
Lecture 5
Discussion 2
Lecture 6
Lecture 7
Lecture 8
Discussion 3
Lecture 9
Lecture 10
Lecture 11
Discussion 4
Lecture 12
Lecture 13
Midterm 1
Lecture 14
Date
Wed Sept 4
Fri Sept 6
Fri Sept 6
Mon Sept 9
Wed Sept 11
Fri Sept 13
Fri Sept 13
Mon Sept 16
Wed Sept 18
Fri Sept 20
Fri Sept 20
Mon Sept 23
Wed Sept 25
Fri Sept 27
Fri Sept 28
Mon Sept 30
Wed Oct 2
TBA
Fri Oct 6
Fri Oct 6
Lecture 15
Lecture 16
Lecture 17
Mon Oct 7
Discussion 5
Fri Oct 12
Wed Oct 9
Fri Oct 11
Mon Oct 14
Lecture 18
Tue Oct 15
Topics
Course overview, motivating examples and course logistics
Sets, experiments, events, probability laws and models
Working with sets
Discrete probability laws and properties of probability laws
Conditional probability laws
Modeling sequential experiments with conditional probability
Analyzing the Monte Hall problem
Total probability theorem and Bayes rule for events
Independence and conditional independence of events
Independent trials and the Binomial law
Analyzing medical testing results
The counting principle and counting permutations
Counting k-permutations and combinations
Counting partitions and applications of counting
Review for Midterm 1
Random variables and probability mass functions
Bernoulli and Binomial random variables
7:00-9:00pm. Covers lectures 1 to 11.
Geometric and Poisson random variables
Discussion section canceled
Functions of random variables and expectations
Higher moments, variance and expected value rule
Properties of expectations, means and variances
Analyzing decision making using expected utility
Holiday – Columbus Day
Multiple random variables: Joint and conditional PMFs
Lecture 19
Lecture 20
Discussion 6
Lecture 21
Lecture 22
Lecture 23
Discussion 7
Lecture 24
Lecture 25
Lecture 26
Discussion 8
Lecture 27
Lecture 28
Midterm 2
Lecture 29
Lecture 37
Mon Dec 2
Multiple random variables: Expectations
Multiple random variables: Independence
Analyzing independence assumptions
Continuous random variables
The Gaussian distribution
Correlation, covariance and causation
Analyzing correlations
Markov inequality
Chebyshev inequality
The weak law of large numbers
Review for Midterm 2
Information, entropy and probability
Compression and coding
Covers lectures 12 to 26.
Error correcting codes
Discussion canceled
Holiday - Veterans' Day
Maximum likelihood inference
Maximum a posteriori inference and priors
Analyzing the effect of priors on inference
Model selection
Model estimation, method of moments and maximum likelihood
Naïve Bayes classification and evidence combination
Classification with Naive Bayes
Estimating naïve Bayes model parameters
General Bayesian networks
Holiday - Thanksgiving
Markov chains: States, transitions and Markov properties
Lecture 38
Wed Dec 4
Markov chains: Path probabilities and n-Step distributions
Lecture 39
Fri Dec 6
Markov chains: Reducibility, periodicity and steady state
Wed Oct 16
Fri Oct 18
Fri Oct 18
Mon Oct 21
Wed Oct 22
Fri Oct 25
Fri Oct 25
Mon Oct 28
Wed Oct 30
Fri Nov 1
Fri Nov 1
Mon Nov 4
Wed Nov 6
TBA
Fri Nov 8
Fri Nov 8
Mon Nov 11
Lecture 30
Lecture 31
Discussion 9
Lecture 32
Lecture 33
Lecture 34
Discussion 10
Lecture 35
Lecture 36
Wed Nov 13
Fri Nov 15
Fri Nov 15
Mon Nov 18
Wed Nov 20
Fri Nov 22
Fri Nov 22
Mon Nov 25
Wed Nov 27
Fri Nov 39
Discussion 11 Fri Dec 6
Review for final exam
Final Exam
Covers lectures 1 to 39
TBA